JP4533603B2 - How to measure skin condition - Google Patents

Description

  The present invention relates to a skin condition discrimination method and a skin condition monitoring method using a discrimination result by the discrimination method.

  The desire for beautiful skin is not only for women but also for everyone, so it can be said that many people are making efforts to keep skin condition favorable by using cosmetics. This skin condition varies greatly from individual to individual, and is a matter that changes with time, such as loss of elasticity and wrinkles, etc., increase with age. Such a change reflects physiological changes of the skin, and it can be said that it is necessary to objectively and appropriately know the state of the skin in order to keep the skin beautiful. Subjectively, the condition of the skin is roughly judged by the person's appearance. For example, whether or not the cosmetics currently used are effective, or whether or not they are on the skin. In order to appropriately determine whether or not the skin condition is appropriate, it is necessary to determine the skin condition so finely that the skin condition can be quantified. From such a need, development of a technique for objectively and appropriately distinguishing the skin condition has been attempted. As such an attempt, for example, a method of capturing the state of the skin as an image with a digital camera or the like, and manipulating the luminance distribution or the like to convert it into a set of numerical values and using it as an index (for example, Patent Document 1, Patent Document 2 and Patent Document 3) and the like, and methods for estimating the skin state from the shape of the stratum corneum cells (for example, refer to Patent Document 4 and Patent Document 5), etc. There is a method of mechanically picking up and digitizing deformation and return (for example, refer to Patent Document 6). However, in these methods, image capture, selection, analysis, etc. are all batch-processed by human hands. It was difficult to say that anyone could easily do this because there were barriers such as the need for them and the preparation and observation of microscope specimens. Or there is a thing which puts a burden on a test subject, and it cannot be denied until now that about wrinkles and skin viscoelasticity, the method of judging visually based on a judgment standard and scoring is the most common.

  On the other hand, in analysis using a near-infrared spectrum, a technique for measuring the near-infrared spectrum of the skin and quantifying the amount of glucose in the blood (see, for example, Patent Document 7 and Patent Document 8) or near-red of the skin A technique for measuring the outer spectrum, extracting a peak derived from water, and quantifying the amount of water in the skin (see, for example, Patent Document 9), and a method for quantifying the amount of subcutaneous fat from the near-infrared spectrum of the skin ( For example, see Patent Document 10) or a method for estimating the severity of a wound from a plurality of spectrum channel images of a wound, wherein a near infrared spectrum image is used as one of the plurality of spectrum channel images. (For example, refer to Patent Document 11), etc., which is a method for distinguishing skin properties, wherein near-infrared absorption spectra of two or more skins having different states are measured in advance, and the near-infrared absorption spectrum A technique for multivariate analysis of the skin property indication value, comparing the result of analysis with the near-infrared absorption spectrum of the skin as the test sample, and distinguishing the skin property of the test sample is as follows: Not known at all. Furthermore, it is already known that statistical analysis methods such as principal component analysis and PLS analysis are applied to spectrum analysis (see, for example, Patent Document 12, Patent Document 13, and Patent Document 14). There is no known technique for quantifying morphological changes from statistical analysis of near-infrared absorption spectra, such as the degree of order of wrinkle or dermal collagen fiber bundle structure in the place of the skin of the chemical composition. Not.

Japanese Patent Laid-Open No. 10-14903 Japanese Patent Laid-Open No. 10-127585 JP 2003-24306 A JP 2001-13138 A JP 09-38045 A Table 01/052724 JP 2003-144421 A JP 2001-37741 A JP 2003-90298 A JP 2000-155091 A Japanese Patent Laid-Open No. 2000-139846 JP 2002-369814 A Japanese National Patent Publication No. 11-502935 Japanese Patent Laid-Open No. 11-142242

  The present invention has been made under such circumstances, and in the differentiation of skin conditions such as wrinkles, skin elastic characteristics, and the order of the dermal collagen fiber bundle structure, anyone can easily make such a differentiation. The issue is to provide.

In view of such circumstances, the present inventors have eagerly sought for a technique by which anyone can easily perform such differentiation in the differentiation of skin conditions such as wrinkles, skin elastic characteristics, and the order of the dermal collagen fiber bundle structure. As a result of repeated research efforts, it is a method for distinguishing skin properties, measuring near-infrared absorption spectra of two or more skins in different states in advance, Multivariate analysis, and using the analysis result as an index, and comparing this with the near-infrared absorption spectrum of the skin that is the test sample, the discrimination of the skin state of the test sample is reproducible and the discriminator We have found that this can be done regardless of the attribute of the present invention, and have completed the invention. That is, the present invention relates to the following technique.
(1) had previously been obtained, and near-infrared absorption spectrum in the wavelength region of 4200～8000Cm ^-1 of two or more skin of different states, the analysis results of the multivariate analysis of the elastic of the skin as an indicator ,
A method of measuring the elasticity of the skin as a test sample from a near-infrared absorption spectrum in the wavelength region of 4200 to 8000 cm ⁻¹ of the skin as the test sample.
(2) The method according to (1), wherein the near infrared absorption spectrum is a Fourier transform near infrared absorption spectrum.
(3) the wavelength region, 4300~5000cm ^-1, 5000~5500cm ^-1, is either 5500~6100cm ^-1, and 6700~7500cm ^-1, the method described in (1) or (2).
(4) Analysis of multivariate analysis of near-infrared absorption spectra in the wavelength region of 4200 to 8000 cm ⁻¹ of two or more types of skin obtained in advance and the score of the order of the dermal collagen fiber bundle structure Using the results as an indicator
A method of calculating an order score of a dermal collagen fiber bundle structure as a test sample from a near-infrared absorption spectrum in a wavelength region of 4200 to 8000 cm ⁻¹ of the skin as a test sample .
(5) The method according to (4), wherein the near infrared absorption spectrum is a Fourier transform near infrared absorption spectrum.
(6) the wavelength region, 4300~5000cm ^-1, 5000~5500cm ^-1, is either 5500~6100cm ^-1, and 6700~7500cm ^-1, the method described in (4) or (5).

  ADVANTAGE OF THE INVENTION According to this invention, in the discrimination of skin conditions, such as a wrinkle, a skin elastic characteristic, and the order of a dermis collagen fiber bundle structure, the technique which anyone can make such a discrimination easily can be provided.

The discrimination method of the present invention is a skin state discrimination method in which near-infrared absorption spectra of skin in two or more different states are measured in advance, and the near-infrared absorption spectrum and the state value of the state are measured. Are analyzed by multivariate analysis, the analysis result is compared with the near-infrared absorption spectrum of the test sample, the state of the test sample is estimated, and this is used as an index. The near-infrared absorption spectrum used for this index or discrimination is a spectrum obtained by a dispersion type using a normal diffraction grating, a spectrum obtained by a device using a diode array, a spectrum obtained by Fourier transforming these, and a detected interferogram. Any spectrum obtained by Fourier transform of can be used. More preferable examples include those obtained by further Fourier transforming the spectrum obtained by the dispersion type apparatus. It is particularly preferable to use a spectrum obtained by performing Fourier transform. Here, multivariate analysis (statistical chemical processing) is used, but multivariate analysis is a relationship between chemical properties such as spectroscopic data and property values such as physical properties. For example, multiple regression analysis or principal component analysis is known. Of these, PLS analysis can be suitably exemplified as the multiple regression analysis. This PLS analysis is based on the difference between a spectral spectrum pattern in which a variable such as absorbance appears and a certain characteristic value of the sample with respect to a continuous change in a factor such as a wavelength in a specific sample. This is an established technique for analyzing the change of each characteristic value and the variable for each factor. In the principal component analysis, the first principal component contributing to the fluctuation is analyzed in the same analysis, and then the second principal component axis orthogonal to the first principal component axis is analyzed. This is a method for comparing and estimating physical properties by pattern changes in the coordinates created by component axes. Multivariate analysis such as PLS analysis or principal component analysis can be performed using commercially available software. As such multivariate analysis software, for example, sold by GL Science, Pirouett, sold by Cybernet Systems, sold by Matlab Yokogawa Electric Co., Ltd. Software such as Simsca (SIMCA) sold by Unscrambler II and Sepanova (SEPANOVA) can be exemplified. In addition to these, an algorithm called SIMCA can be added. Such an algorithm is often incorporated in the software and is useful for displaying principal component analysis. When these softwares are used to analyze near-infrared absorption spectra and the results are used in the discrimination method of the present invention, the general processing steps are as follows. At this time, the Fourier transform near-infrared absorption spectrum to be used may be an original spectrum obtained by measurement, or may be obtained by processing the original spectrum. Data processing methods include, for example, first-order differential values, second-order differential values, third-order differential values, etc., smoothing (Smoothing), normalization (Nor
Preferred examples include malize), MSC (Multiplicative Scatter Correction), SNV (Standard Normal Variate), averaging (Mean-Center), and autoscale. Of these, the original spectrum or its second derivative is preferred. Thus, when analyzed, there is a good correlation between the readings representing the skin condition and the Fourier transform near infrared absorption spectrum of the skin.

  On the other hand, factors commonly used in the cosmetics and dermatological fields can be used as the skin condition. Such factors include, for example, surface morphological factors such as wrinkles, rough skin, roughness, and a feeling of firmness, skin elasticity, skin viscoelastic properties such as beams, order of dermal collagen fiber bundle structure, stratum corneum structure, etc. Physiological characteristics such as skin structure characteristics, transdermal water dispersion loss (TEWL), lipid metabolism and the like can be suitably exemplified. Among these, preferred is a highly effective degree of wrinkle, saying that “an evaluation technique that is difficult to quantify or a technique that requires time and skill to learn is objective and easy to quantify”. The degree of order of skin viscoelasticity and dermal collagen fiber bundle structure can be particularly preferably exemplified. The degree of wrinkles can be used by processing a replica image by a computer using an imaging unit and digitizing the replica. As a value, a wrinkle area ratio and a wrinkle volume ratio are obtained. The order of the dermal collagen fiber bundle structure can be determined by an expert's visual judgment, for example, a score value, and the skin viscoelasticity is obtained by sucking the skin using a measuring device called a “cutmeter”. The degree of deformation of the skin by the suction and the degree of return of the deformation can be converted into numerical values. For more accurate measurement, a representative value of skin deformation characteristics can be used with a measuring instrument called “resiliometer” described in Patent Document 6. Regarding the dermal collagen fiber bundle structure, the value obtained by scoring the electron microscopic image of the skin collected from humans, or by irradiating the animal with ultraviolet rays for a long period of time, artificially destroys the dermal collagen fiber bundle structure. It is also possible to use visual judgment.

As a Fourier transform near infrared absorption spectrum used in the discrimination method of the present invention, at least 400 cm-1 of 4000 to 8000 cm-1 is a preferable wavelength region, and in a particularly preferable wavelength region, 4300 to 5000 cm-1, 5000. -5500 cm-1, 5500-6100 cm-1, and 6700-7500 cm-1. This is because the spectrum in this wavelength region well reflects the characteristic value of the skin condition. The near-infrared absorption spectrum in this range is considered to be due to the fact that the existence state and behavior of proteins in the skin are accurately captured.

  Thus, the measured near-infrared absorption spectrum is preferably subjected to a Fourier transform and then subjected to statistical chemical analysis together with the skin condition indication value, and the causal relationship thereof is quantified. The skin condition of the test sample is discriminated from the comparison between the quantity relationship and the spectrum of the test sample. These specific procedures are shown below.

In the case of PLS analysis (1) The near-infrared absorption spectrum of the skin dispersion type or diode array type, or their Fourier transform spectrum or Fourier transform spectrum, if desired, data processing such as second derivative, wavelength and near infrared absorption A matrix with the spectrum or the processed data is created.
(2) A matrix of the matrix and the characteristic value is created, a near-infrared absorption spectrum or its processed data having a large movement is extracted with respect to the movement of the characteristic value, and the wavelength is specified. (3) A calibration curve is created from the extracted near-infrared absorption spectrum or the processed data and the characteristic value. At the same time, a plot on the calibration curve is created for each characteristic value.
(4) The Fourier transform near-infrared absorption spectrum of the test sample is measured, and data such as second derivative is processed as desired.
(5) Extract the data of the wavelength specified in (2) from the data of (4).
(6) Create a map of the data extracted in (5) onto the calibration curve. Alternatively, the data is plotted on a calibration curve.
(7) The plot for each characteristic value in (3) is compared with the mapping or plot in (5) to estimate the characteristic value of the sample.
(2) The following operations can be performed using computer software.

In the case of principal component analysis (1) The near-infrared absorption spectrum of the skin dispersion type or diode array type, or the Fourier transform spectrum or Fourier transform spectrum thereof, if desired, is subjected to data processing such as second derivative, and the wavelength and near infrared An absorption spectrum or a matrix with the processed data is created.
(2) A principal component analysis is performed on the matrix to create a first principal component axis.
(3) Create a second principal component axis orthogonal to the first principal component.
(4) The points formed by the first principal component and the second principal component of the spectrum of (1) are plotted on the plane formed by the first principal component axis and the second principal component axis.
(5) Grouping is performed using an algorithm such as shimuka as desired.
(6) The near-infrared spectrum of the test sample is measured in the same manner as (1), and the same plot as in (4) is performed.
(7) The test sample is identified using the plot in (4) or the grouping in (5) as an index.

The discrimination method of the present invention can also be used to group skin conditions and select cosmetics or aesthetic courses suitable for the skin condition for selection of cosmetics or courses such as esthetics. In addition, the effect of treatment with cosmetics or treatment with esthetics can be used for skin condition monitoring by distinguishing the skin condition over time and tracing the change. Or, identify the internal state of the skin and predict future effects on the surface morphology,
It can also be used to select cosmetics or aesthetic courses suitable for the internal skin condition.

  Hereinafter, the present invention will be described in more detail with reference to examples, but it is needless to say that the present invention is not limited to such examples.

<Reference example>
Preparation of model animal According to the schedule shown in FIG. 42, the back of a hairless mouse (female, 5 weeks old) was irradiated with ultraviolet rays of 54 to 108 mJ / cm 2 once a day, 3 times a week, gradually increasing the amount of irradiation energy. Irradiation was continued for 2 to 10 weeks, and photoaging with different degrees for each group was caused on the skin, and this was used as a model animal.

<Example 1>
The Fourier transform near infrared absorption spectrum of the back of the animal of the reference example was measured. Thereafter, the skin elasticity was measured with a cutometer, and the order of the dermal collagen fiber bundle structure was visually determined. In the cute meta, a parameter (see FIG. 1) represented by Ur * of the curve drawn by this was used. After this measurement, the order of the dermal collagen fiber bundle structure was evaluated by evaluating the skin by collecting skin from an animal, preparing a specimen, and scoring under an electron microscope according to the following criteria. Score 3: Strong dermal collagen fibers are observed in the entire visual field. Score 2: Strong dermal collagen fibers are observed in more than half of the visual field. Slightly broken fiber bundle structure is observed. Score 1: Slightly robust dermal collagen fibers are observed. No microscopic fiber bundle structure is observed, score 0: a criterion that no robust dermal collagen fibers are observed. In addition, the thickness of the epidermis was measured. For the volume ratio of wrinkles in the epidermis, a replica was taken from the back of the animal, the replica was subjected to image processing by a computer using an imaging unit, and used as a numerical value. Using these values and the near infrared absorption spectrum, PLS analysis was performed using Unscrambler II, and a calibration curve was prepared by PLS analysis. The calibration curves are shown in FIGS. 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30, 32, 34, 36, 38, 40. From this, it can be seen that there is a very good correlation between the near-infrared absorption spectrum and the skin condition indication value. Furthermore, these principal component analyzes were performed using Unscrambler II. The results are shown in FIGS. 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31, 33, 35, 37, 39, 41. From this, it can be seen that the plots are grouped for each characteristic value to form a group. If it is known what the attribute of this group is, it can be understood from the near-infrared absorption spectrum that the attribute of the test sample becomes clear. In addition, it can be clearly seen that the discrimination method of the present invention has an advantage that discrimination can be instantly performed for a plurality of characteristic values.
Near-infrared spectrometer: VECTER 22 / N (Bruker Optics)
Replica imaging unit: ASA-03R-U

  Since the present invention can distinguish the internal structure of the skin non-invasively with almost no burden on the subject, it can provide skin data for selecting an appropriate cosmetic without pain and can provide cosmetic sales support counseling. Available for tools.

It is a figure which shows the parameter represented by Ur * of a cut meter. It is a figure which shows the correlation of the wrinkle volume of Example 1, and a spectrum. (Wavelength 4300-5000cm-1) It is a figure which shows the result of the principal component analysis of the wrinkle volume of Example 1, and a spectrum. (Wavelength 4300-5000cm-1) It is a figure which shows the correlation of the wrinkle volume of Example 1, and a spectrum. (Wavelength 5000-5500cm-1) It is a figure which shows the result of the principal component analysis of the wrinkle volume of Example 1, and a spectrum. (Wavelength 5000-5500cm-1) It is a figure which shows the correlation of the wrinkle volume of Example 1, and a spectrum. (Wavelength 5500-6100 cm-1) It is a figure which shows the result of the principal component analysis of the wrinkle volume of Example 1, and a spectrum. (Wavelength 5500-6100 cm-1) It is a figure which shows the correlation of the wrinkle volume of Example 1, and a spectrum. (Wavelength 6700-7500cm-1) It is a figure which shows the result of the principal component analysis of the wrinkle volume of Example 1, and a spectrum. (Wavelength 6700-7500cm-1) It is a figure which shows the correlation of the wrinkle volume of Example 1, and a spectrum. (Wavelength 4200-8000 cm-1) It is a figure which shows the result of the principal component analysis of the wrinkle volume of Example 1, and a spectrum. (Wavelength 4200-8000 cm-1) It is a figure which shows correlation of Ur * of Example 1 and a spectrum. (Wavelength 4300-5000cm-1) It is a figure which shows the result of the principal component analysis of Ur * and a spectrum of Example 1. (Wavelength 4300-5000cm-1) It is a figure which shows correlation of Ur * of Example 1 and a spectrum. (Wavelength 5000-5500cm-1) It is a figure which shows the result of the principal component analysis of Ur * and a spectrum of Example 1. (Wavelength 5000-5500cm-1) It is a figure which shows correlation of Ur * of Example 1 and a spectrum. (Wavelength 5500-6100 cm-1) It is a figure which shows the result of the principal component analysis of Ur * and a spectrum of Example 1. (Wavelength 5500-6100 cm-1) It is a figure which shows correlation of Ur * of Example 1 and a spectrum. (Wavelength 6700-7500cm-1) It is a figure which shows the result of the principal component analysis of Ur * and a spectrum of Example 1. (Wavelength 6700-7500cm-1) It is a figure which shows correlation of Ur * of Example 1 and a spectrum. (Wavelength 4200-8000 cm-1) It is a figure which shows the result of the principal component analysis of Ur * and a spectrum of Example 1. (Wavelength 4200-8000 cm-1) It is a figure which shows the correlation of the score and spectrum of the order of the dermal collagen fiber bundle of Example 1. (Wavelength 4300-5000cm-1) It is a figure which shows the result of the principal score analysis of the score of the dermis collagen fiber bundle of Example 1, and a spectrum. (Wavelength 4300-5000cm-1) It is a figure which shows the correlation of the score and spectrum of the order of the dermal collagen fiber bundle of Example 1. (Wavelength 5000-5500cm-1) It is a figure which shows the result of the principal score analysis of the score of the dermis collagen fiber bundle of Example 1, and a spectrum. (Wavelength 5000-5500cm-1) It is a figure which shows the correlation of the score and spectrum of the order of the dermal collagen fiber bundle of Example 1. (Wavelength 5500-6100 cm-1) It is a figure which shows the result of the principal score analysis of the score of the dermis collagen fiber bundle of Example 1, and a spectrum. (Wavelength 5500-6100 cm-1) It is a figure which shows the correlation of the score and spectrum of the order of the dermal collagen fiber bundle of Example 1. (Wavelength 6700-7500cm-1) It is a figure which shows the result of the principal score analysis of the score of the dermis collagen fiber bundle of Example 1, and a spectrum. (Wavelength 6700-7500cm-1) It is a figure which shows the correlation of the score and spectrum of the order of the dermal collagen fiber bundle of Example 1. (Wavelength 4200-8000 cm-1) It is a figure which shows the result of the principal score analysis of the score of the dermis collagen fiber bundle of Example 1, and a spectrum. (Wavelength 4200-8000 cm-1) It is a figure which shows the correlation of the skin thickness of Example 1, and a spectrum. (Wavelength 4300-5000cm-1) It is a figure which shows the result of the principal component analysis of the skin thickness of Example 1, and a spectrum. (Wavelength 4300-5000cm-1) It is a figure which shows the correlation of the skin thickness of Example 1, and a spectrum. (Wavelength 5000-5500cm-1) It is a figure which shows the result of the principal component analysis of the skin thickness of Example 1, and a spectrum. (Wavelength 5000-5500cm-1) It is a figure which shows the correlation of the skin thickness of Example 1, and a spectrum. (Wavelength 5500-6100 cm-1) It is a figure which shows the result of the principal component analysis of the skin thickness of Example 1, and a spectrum. (Wavelength 5500-6100 cm-1) It is a figure which shows the correlation of the skin thickness of Example 1, and a spectrum. (Wavelength 6700-7500cm-1) It is a figure which shows the result of the principal component analysis of the skin thickness of Example 1, and a spectrum. (Wavelength 6700-7500cm-1) It is a figure which shows the correlation of the skin thickness of Example 1, and a spectrum. (Wavelength 4200-8000 cm-1) It is a figure which shows the result of the principal component analysis of the skin thickness of Example 1, and a spectrum. (Wavelength 4200-8000 cm-1) It is drawing which described the preparation schedule of the photoaging animal model of a reference example.

Claims (6)

Had previously been obtained, and near-infrared absorption spectrum in the wavelength region of 4200～8000Cm ^-1 of two or more skin of different states, the analysis results of the multivariate analysis of the elastic of the skin as an indicator,
A method of measuring the elasticity of the skin as a test sample from a near-infrared absorption spectrum in the wavelength region of 4200 to 8000 cm ⁻¹ of the skin as the test sample. The method according to claim 1, wherein the near infrared absorption spectrum is a Fourier transform near infrared absorption spectrum. The method according to claim ¹ , wherein the wavelength region is any of 4300 to 5000 cm ⁻¹ , 5000 to 5500 cm ⁻¹ , 5500 to 6100 cm ⁻¹ , and 6700 to 7500 cm ⁻¹ . 4200 to 8000 cm of two or more kinds of skins obtained in advance. ^-1 Using the analysis results of multivariate analysis of the near-infrared absorption spectrum in the wavelength region of the skin and the order score of the dermal collagen fiber bundle structure as an index,
4200-8000 cm of skin as test sample ^-1 The score of the order of the dermal collagen fiber bundle structure which is the said test sample is calculated from the near-infrared absorption spectrum of the wavelength region. The method according to claim 4, wherein the near infrared absorption spectrum is a Fourier transform near infrared absorption spectrum. The wavelength region is 4300 to 5000 cm ^-1 , 5000-5500cm ^-1 5500-6100cm ^-1 And 6700-7500cm ^-1 The method according to claim 4 or 5, which is either

Images